Reports of increasing prevalence of autism spectrum disorders (ASDs), a set of highly genetic conditions, are intensifying interest in the role of environmental exposures, including infectious, immune, and toxic factors. Retrospective studies exploring associations between environmental factors and ASDs are compromised by selection bias, small sample sizes, possibly invalid diagnosis, and absence of biologic measures. This prospective study will employ an unselected birth cohort of 75,500 in which cases are ascertained through screening of the entire population, diagnoses established using uniform procedures, extensive histories and clinical data obtained, and biologic samples collected serially throughout pregnancy and early childhood. The application of high throughput laboratory assays to derive maximal information from developmentally-influenced, finite, and nonrenewable biologic samples, and inclusion of early screening and diagnostic assessments, will permit an unprecedented, rich view of the longitudinal trajectory and nascent signs and symptoms of ASDs, facilitate discovery of biomarkers, and afford unique insights into the role of gene:environment interactions in ASD pathogenesis. Specific aims are to: (1) establish the Autism Birth Cohort (ABC) through ascertainment of cases of autism spectrum disorder (ASD, N=150-233) and selection of controls (N-1000) from the Norway Mothers and Child (MoBa) cohort; (2) examine biologic pathways that may predispose to ASD, through evaluation of immune, endocrine, and neuroregulatory factors in mothers during early gestation or at birth and in children, at birth or 30 months postnatal; (3) identify environmental factors that may be directly or indirectly associated with ASD, including pre- or postnatal infection, vaccination, very low birth weight or other obstetric risk factors in which infections are implicated, dietary and/or environmental exposure to methylmercury; (4) describe the natural history of clinical, anthropometric, and neurobehavioral features of ASD; and (5) explore genotypic influences that may be directly or indirectly associated with ASD by testing associations of ASD and/or its endophenotypes with family history of autoimmune disease or selected candidate genes, and investigating conditional gene-environment effects using antecedent factors found to influence ASD risk.